Katerelos v. Commissioner

HARALAMPOS KATERELOS AND IRENE KATERELOS, Petitioners v. COMMISSIONER OF INTERNAL REVENUE, Respondent Katerelos v. Commissioner Docket No. 15202-94 United States Tax Court T.C. Memo 1996-340; 1996 Tax Ct. Memo LEXIS 355; 72 T.C.M. (CCH) 205; July 29, 1996, Filed *355 Decision will be entered under Rule 155. Stefano D. Corradini, for petitioners. John R. Mikalchus, for respondent. CHIECHI CHIECHI MEMORANDUM FINDINGS OF FACT AND OPINION CHIECHI, Judge: Respondent determined the following deficiencies in, additions to, and penalties on petitioners' Federal income tax: Section Section Section Section Year Deficiency 6653(b)(1)(A) 1 6653(b)(1)(B) 6653(b)(1) 6663(a) [1987] $ 23,437 $ 10,037 * -- -- [1988] 25,077 -- -- $ 13,613 -- [1989] 16,082 -- -- -- $ 8,189 [1990] 25,676 -- -- -- 16,498 [1991] 4,021 -- -- -- -- * 50 percent of the interest due on the portion of the underpay- ment attributable to fraud. Respondent determined that $ 13,383 of the underpayment for 1987 was attributable to fraud. The issues remaining for decision are: (1) Is petitioner Haralampos Katerelos (Mr. Katerelos) liable for 1987 and 1988 for the additions to tax for fraud under section 6653(b)(1) and for 1989 and 1990 for the penalties for*356 fraud under section 6663(a) on the portion of the underpayment for each of those years that is attributable to his not reporting certain gross receipts in petitioners' Federal income tax return (return) for each of those years? We hold that he is. (2) Did petitioners have gross receipts for each of the years 1987 through 1990 in excess of those reported in Schedule C of their return for each of those years? We hold that they did. (3) Are petitioners entitled to deductions under section 162(a) for automobile expenses for each of the years at issue? We hold that they are not. (4) Did petitioners place depreciable properties in service during 1986 with bases in excess of $ 22,102 to which the parties stipulated? We hold that they did not. (5) Are petitioners entitled to a general business credit carryforward from 1986? We hold that they are not. FINDINGS OF FACT 2 *357 Some of the facts have been stipulated and are so found. Petitioners resided in Tucson, Arizona, at the time the petition was filed. Petitioners, who emigrated to the United States from Greece, have three children, all of whom they claimed as dependents in the returns that they filed for the years at issue. Petitioners' Restaurant Business--New Dolce Vita In General Petitioners have been involved in the restaurant business in the United States since 1977. In December 1985, petitioners executed a lease for property located at 6781 North Thornydale in Tucson, Arizona, at which they planned to operate a restaurant called New Dolce Vita (NDV) that was to specialize in Italian, American, and Greek cuisine. Petitioners provided all the fixtures and equipment for the restaurant, some of which they purchased used, and not new. One of the items that petitioners furnished NDV was a Casio cash register (cash register) that they purchased during 1986 from Cactus King Cash Register Co. (Cactus King). The cash register functioned properly throughout the years at issue. Petitioners opened NDV for business in April 1986 and operated it as a sole proprietorship throughout the years*358 at issue. During those years, NDV served breakfast, lunch, and dinner, seven days a week, from approximately 7 a.m. to about 10 p.m., except on approximately five days during each such year when it was closed for certain holidays. Throughout the years at issue, Mr. Katerelos, who was responsible for the management of NDV, worked approximately 105 hours per week at that restaurant and was involved in every aspect of its operations, including mopping the floors, washing the dishes, cooking the food, and purchasing the supplies. During those years, petitioner Irene Katerelos (Ms. Katerelos), petitioners' daughter Anneta, and Mr. Katerelos' brother Evangelos also worked at NDV. During the years at issue, NDV's highest sales volumes generally occurred on Friday and Saturday evenings and on Sunday mornings and afternoons. It also experienced large sales volumes on special days such as Mother's Day and Valentine's Day. The winter months were the busiest months for NDV because of the high number of visitors to Tucson during that time of year. During the years at issue, petitioners employed two to six servers for the breakfast and lunch shifts on weekdays. On weekday evenings and on weekends*359 during those years, petitioners employed a hostess and two to six servers. Cash Register at NDV At the time petitioners purchased the cash register in 1986, Tony Mastrangelo (Mr. Mastrangelo), the owner of Cactus King, trained Mr. Katerelos in its operation. 3 During the years at issue, when a hostess was on duty, that person generally operated the cash register and was responsible for entering sales into it, and when no hostess was on duty, or if the hostess was busy, the servers entered their own sales into the cash register. During those years, Mr. Katerelos also entered sales into the cash register. 4 *360 Operation of the Cash Register [5] The cash register could have been set at one of the following different modes by turning a key in the mode switch: (1) Off, (2) register, (3) refund, (4) Z or reset, and (5) X or read. For purposes of entering sales into the cash register, the mode switch was set to the register mode. The operator of the cash register entered a sale into the machine by entering the price of an item and pressing one of the various buttons known and hereinafter referred to as department buttons. The cash register was programmed so that the department buttons could be used to distinguish among breakfast, lunch, and dinner sales. 6 The entry of a transaction into the wrong department did not affect the total amounts of gross receipts of NDV that Mr. Katerelos was required to report in petitioners' returns for the years at issue. *361 After the operator finished entering all of the items for a particular transaction into the cash register, the operator could press a button known and hereinafter referred to as the subtotal button. When pressed down, the subtotal button added all of the items entered as part of a transaction to arrive at a subtotal for the amount of the transaction, calculated the sales tax due, and added the sales tax to that subtotal to arrive at the total for that transaction (transaction total). After pressing the subtotal button, the operator completed the transaction and opened the cash register drawer by (1) entering the amount of cash tendered by the customer into the cash register and pressing a button known and hereinafter referred to as the cash/amount tendered button or (2) pressing a button known and hereinafter referred to as the charge button if the customer was paying by credit card. Instead of pressing the subtotal button and either the cash/amount tendered button or the charge button, the operator could skip the step of pressing the subtotal button by pressing the cash/amount tendered button or the charge button, which caused the cash register to calculate the transaction total*362 for the transaction and open the cash register drawer in one step, rather than two steps. Although the cash register was equipped to distinguish between cash and credit sales, Mr. Katerelos did not distinguish between those two types of sales when he was operating the cash register. 7 Failure to distinguish between cash and credit sales did not affect the total amounts of gross receipts of NDV that Mr. Katerelos was required to report in petitioners' returns for the years at issue. The cash register was programmed to maintain certain data or information in its memory in order to facilitate calculation of the sales of a business. Activation of the Z or reset mode on the cash register by turning the mode switch to that mode and pressing the cash/amount tendered button accessed that information. Mr. Katerelos was the only person who activated the Z or reset mode on the cash register*363 during the years at issue. 8 Activation of the Z mode (1) caused the cash register to print on a cash register tape 9*364 a summary of all the transactions that occurred since the last activation of the Z mode and (2) reset all amounts to zero other than the grand total as of the time of the Z reading and the transaction number as of the time of the Z reading, both of which were not resettable. The data printed on the cash register tape upon activation of the Z mode (known and hereinafter referred to as a Z reading) included, inter alia, the letter "Z", the date of the Z reading, the total of all transaction totals since the last Z reading (total sales), the nonresettable grand total as of the time of the Z reading, and the nonresettable transaction number as of the time of the Z reading. 10 Throughout the period January 1, 1987, through May 29, 1990, the nonresettable grand total as of the time of a Z reading (grand total) equaled the sum of all transaction totals entered into the cash register that commenced with the time at which the cash register was first used in 1986 and that ended with the time of that Z reading. 11 From May 30, 1990, through December 31, 1990, the nonresettable grand total as of the time of a Z reading equaled the sum of all transaction totals entered into the cash register that commenced on May 30, 1990, and that ended with the time of that Z reading. *365 Each time a transaction was entered into the cash register, the grand total increased by that transaction total. The grand total could have included transactions that did not constitute sales (nonsale transactions), such as transactions entered while programming the cash register. A person could not have simply entered a number into the cash register as a grand total. The only way to reset the grand total to zero was to cause the capacitors to drain by unsoldering and taking the battery out of circuit. In addition to maintaining the grand total, *366 the cash register maintained a nonresettable, consecutive transaction number for every transaction entered into it (transaction number). 12 The cash register automatically assigned a transaction number to each transaction entered into it that increased by one each time a new transaction was entered into the cash register and printed that assigned transaction number on the cash register tape as part of that transaction. Opening the cash register by pressing a button known and hereinafter referred to as the no-sale button or activating the Z or reset mode also caused the cash register to record a transaction and to increase the transaction number by one. For the periods January 1, 1987, through September 14, 1987, and May 30, 1990, through December 31, 1990, in addition to the functions described above, the cash register was programmed to print on the cash register tapes the time of day of every transaction*367 using a 24-hour clock (clock). Cash Register Tapes of NDV Saved by Mr. Katerelos During the years at issue, Mr. Katerelos saved either the entire cash register tape for each day on which NDV was open for business or only a portion of each such tape. Each such tape or portion thereof showed, inter alia, all the transaction totals since the last activation of the Z or reset mode, and the last item printed thereon was a Z reading. For the period January 1987 through about September 1990, Mr. Katerelos' decision as to whether to save the entire cash register tape or only a portion thereof for each day on which NDV was open for business frequently depended on whether he had activated the Z or reset mode more than once on a particular day. The cash register tapes that Mr. Katerelos saved for a minimum of 650 days during that period on which he had activated the Z mode more than once daily consisted of only a portion of the entire cash register tape for each such day, and that portion showed only one, and not all, of the Z readings for the day. 13 Mr. Katerelos discarded the remaining portion of the cash register tape for each such day. (Unless otherwise indicated, hereinafter, we*368 shall refer to the portions of the cash register tapes that Mr. Katerelos discarded as the missing portions of the cash register tapes, and references to the cash register tapes that Mr. Katerelos saved shall be to the portions of those tapes he saved.) In addition to discarding the missing portions of the cash register tapes, Mr. Katerelos also discarded all of the individual guest checks for the food and beverages sold during the years at issue (guest checks). Mr. Katerelos was aware during the years at issue that he was required to report as gross receipts in petitioners' returns for those years the receipts from all sales made by NDV during those years. The amount of gross receipts of NDV that petitioners reported in their return for each of the years at issue was based upon the amount, if any, of gross receipts of NDV that Mr. Katerelos had recorded for each day during those*369 years in a gross receipts journal that he maintained for NDV for each of those years (gross receipts journals). The amount of daily gross receipts of NDV that Mr. Katerelos recorded in those journals during the years at issue was based upon the amount of the total sales shown as part of the Z reading on the cash register tape that he saved for each day during those years. Since the cash register did not have a cancel function, if the operator made a mistake in entering the amount of any item for a particular transaction so that the transaction total for that transaction also was incorrect, the operator had to start over and correctly reenter into the cash register all items that were part of that transaction so that the transaction total for that transaction was correct. Those corrected items were reentered as items for a new transaction that had a correct transaction total. Hereinafter, we shall refer to transactions with incorrect transaction totals due to errors made by the operators of the cash register in entering transactions into it as overrings. During the years at issue, Mr. Katerelos and/or other operators of the cash register generally kept track of overrings by circling*370 any overring transaction on the cash register tape. However, the transaction total of every transaction entered into the cash register for a day, including any overring transaction, was reflected in the total sales for that day. 14 To correct that problem, when a transaction was circled as an overring on the cash register tape that Mr. Katerelos saved for a particular day, Mr. Katerelos subtracted the transaction total of any such transaction from the amount of daily total sales according to the Z reading on the cash register tape that he saved for that day and entered the balance as the amount of NDV's daily gross receipts that he recorded for that day in the gross receipts journal. *371 For approximately 10 days during the years 1987 through 1990, Mr. Katerelos saved cash register tapes that showed two Z readings. The first Z reading on such a tape resulted from Mr. Katerelos' having activated the Z or reset mode after an error (that may or may not have been an overring) had occurred during the day in entering a transaction into the cash register. After having generated the first Z reading on such a tape, Mr. Katerelos correctly reentered all the transactions that had thus far occurred for the day 15 and entered additional transactions as they occurred during that day. Thereafter, on that same day, Mr. Katerelos activated the Z mode a second time to generate the second Z reading on the cash register tape that he saved for the day. Mr. Katerelos recorded in the gross receipts journals the total sales from the second Z reading for the day as the amount of NDV's gross receipts for that day. 16 *372 During the years at issue, Irwin Berlat (Mr. Berlat), petitioners' accountant, 17 used the daily amount of gross receipts of NDV that Mr. Katerelos had recorded in the gross receipts journals for those years in order to calculate the amounts of NDV's monthly 18 and annual gross receipts for those years. In addition to giving Mr. Berlat the gross receipts journals that Mr. Katerelos maintained for the years at issue, Mr. Katerelos provided him with the daily cash register tapes that he had saved for those years and that he had used to arrive at the amounts of daily gross receipts he recorded in those journals. Mr. Berlat used those gross receipts journals to prepare petitioners' returns for the years 1987 through 1989, and another accountant, Don Bailey (Mr. Bailey), used those journals to prepare their returns for 1990 and 1991. *373 For certain days during the years at issue, there is a gap between the last transaction number shown on the cash register tape that Mr. Katerelos saved for a particular day and the first transaction number shown on the cash register tape that he saved for the next day. Neither those missing transaction numbers nor the transactions to which the cash register assigned those numbers appear on any of the cash register tapes that Mr. Katerelos saved for the years at issue. (We shall sometimes refer (1) to the transactions that were entered into the cash register during the years at issue but that do not appear on any of the cash register tapes Mr. Katerelos saved for those years as the missing transactions and (2) to the transaction numbers assigned to those missing transactions as the missing transaction numbers.) The missing transactions included transactions that increased the grand total maintained by the cash register. 19 *374 To illustrate the foregoing, the last transaction number shown on the cash register tape that Mr. Katerelos saved for January 3, 1987, was 2,886, and the grand total according to the Z reading on that tape was $ 246,182.93. The first transaction number shown on the cash register tape that he saved for the next day, January 4, 1987, was 2,966, and the grand total and the total sales according to the Z reading on that tape were $ 247,681.04 and $ 689.48, respectively. Thus, 79 transactions that were assigned 79 transaction numbers were entered into the cash register between the last transaction number shown on the cash register tape that Mr. Katerelos saved for January 3, 1987, and the first transaction number shown on the cash register tape that he saved for January 4, 1987. Neither those 79 missing transactions nor the 79 missing transaction numbers assigned to those transactions appear on any of the cash register tapes that Mr. Katerelos saved for the years at issue. In addition, the difference between the grand totals according to the respective Z readings on the cash register tapes that Mr. Katerelos saved for those two days equaled $ 1,498.11. However, the total sales according*375 to the Z reading on the cash register tape that he saved for January 4, 1987, as well as the amount of gross receipts of NDV for that day that he recorded in the gross receipts journal for 1987, was only $ 689.48. The difference between the respective grand totals for those two days (viz., $ 1,498.11) exceeded the total sales according to the Z reading on the cash register tape that Mr. Katerelos saved for January 4, 1987 (viz., $ 689.48) by $ 808.63. Such $ 808.63 excess was the total of the various amounts entered into the cash register for the 79 missing transactions between January 3, 1987, and January 4, 1987. Respondent's Audit During the summer of 1990, respondent's revenue agent, James Phielix (Mr. Phielix), conducted an audit of petitioners' returns for 1987, 1988, and 1989. Mr. Phielix' examination of their 1990 and 1991 returns did not commence until early 1993 because of an ongoing criminal tax investigation of Mr. Katerelos that ended when Mr. Katerelos was informed by correspondence dated December 15, 1992, that he was no longer the subject of such an investigation. During the course of his examination of petitioners' returns for the years at issue, Mr. Phielix*376 created a handwritten summary that showed for each day during the years at issue the following data or information displayed on the cash register tape that Mr. Katerelos saved for each day during those years: The date, the total sales according to the Z reading, the first and last transaction numbers, the number of missing transaction numbers between the first transaction number on the cash register for each day and the last transaction number on the cash register tape for the previous day, and the nonresettable grand total according to the Z reading. 20 That summary also showed the difference that Mr. Phielix had calculated between the grand total according to the Z reading for each day that was displayed on each of those tapes and the grand total according to the Z reading for the previous day that was displayed on each of those tapes. (Hereinafter, we shall sometimes refer to that difference as the Z reading difference in grand totals. 21) Mr. Phielix transferred the information for 1987 through 1990 that was contained in his handwritten summary (with certain changes 22*378 ) into a computer from which he produced a computerized summary that displayed the same information as the *377 handwritten summary (first computerized summary). 23 In addition to the first computerized summary, Mr. Phielix created a second computerized summary (second computerized summary) to display the amount of additional gross receipts of NDV that he had calculated for each day during 1987 through 1990 on which NDV was open for business. Mr. Phielix used the information in the first computerized summary to determine whether the gross receipts of NDV were underreported in petitioners' returns for 1987 through 1990. In that connection, Mr. Phielix treated the amount of any Z reading difference in grand totals during the years 1987 through 1990 as the amount of gross receipts of NDV that petitioners were required to report for the last day of the multiday period in respect of which Mr. Phielix had compared the respective grand totals displayed on the cash register tapes that Mr. Katerelos saved for those days. To illustrate, *379 Mr. Phielix treated the $ 1,498.11 Z reading difference in grand totals that were shown on the respective January 4, 1987 and January 3, 1987 cash register tapes that Mr. Katerelos saved for those days as the amount of NDV's gross receipts for January 4, 1987. In determining the amount of gross receipts that petitioners were required to report for a particular day and that Mr. Phielix displayed in the second computerized summary, Mr. Phielix made certain adjustments to the Z reading differences in grand totals that he had calculated by using the cash register tapes that Mr. Katerelos saved. For any day on which the cash register tape that Mr. Katerelos saved showed that an overring occurred, Mr. Phielix subtracted the amount of any such indicated overring from the amount of gross receipts of NDV that he had calculated for that day on the basis of the Z reading difference in grand totals. For any day on which the cash register tape that Mr. Katerelos saved showed two Z readings, Mr. Phielix treated the total sales according to the first Z reading shown on that tape as an overring and subtracted the amount of those total sales from the amount of gross receipts of NDV that he had calculated*380 for that day on the basis of the Z reading difference in grand totals by using the second Z reading on that tape. 24 For any day on which the cash register tape that Mr. Katerelos saved showed that there was a cash shortage in the cash register for that day, Mr. Phielix subtracted the amount of any such shortage from the amount of gross receipts of NDV that he had calculated for that day on the basis of the Z reading difference in grand totals. *381 In determining the amounts of gross receipts of NDV, if any, that petitioners failed to report in their returns for the years 1987 through 1990 and that Mr. Phielix displayed in the second computerized summary, Mr. Phielix subtracted the amount of NDV's gross receipts that Mr. Katerelos had recorded in the gross receipts journals for each day during the years 1987 through 1990 from the amount of gross receipts of NDV that he had calculated for each such day on the basis of the Z reading difference in grand totals (adjusted as described above by any amount treated as an overring or cash shortage). 25 Mr. Phielix treated any resulting difference as additional gross receipts of NDV for that day. *382 The second computerized summary displayed the following information for each day during 1987 through 1990 on which NDV was open for business: (1) The number of transactions shown on the cash register tape that Mr. Katerelos saved for each such day, (2) the amount of gross receipts of NDV that Mr. Katerelos recorded in the gross receipts journals for each such day, (3) the number of missing transaction numbers between the first transaction number on the cash register tape that Mr. Katerelos saved for each such day and the last transaction number on the cash register tape that he saved for the preceding day, (4) the amount of gross receipts of NDV that Mr. Phielix had calculated for each such day on the basis of the Z reading difference in grand totals (adjusted as described above by any amount treated as an overring or cash shortage), (5) the amount, if any, treated as an overring or cash shortage for each such day, and (6) the amount, if any, of additional gross receipts of NDV that Mr. Phielix had calculated for each such day (or multiday period). The second computerized summary showed that the amount of daily gross receipts of NDV that Mr. Katerelos had recorded in the gross receipts*383 journals for a minimum of 650 of the 1,440 days throughout the period 1987 through 1990 on which NDV was open for business was less than the amount of daily gross receipts of NDV that Mr. Phielix had calculated for each such day on the basis of the Z reading difference in grand totals. The second computerized summary further showed that, throughout the period 1987 through 1990, over 11,400 of the approximately 140,000 transactions that were entered into the cash register during that period do not appear on any of the cash register tapes that Mr. Katerelos saved for those years. That summary also showed that the total of the various transaction totals entered into the cash register for those missing transactions during the period that began on January 1, 1987, and that ended on December 31, 1990, was approximately $ 199,000. In order to determine how many hours NDV was open according to the cash register tape that Mr. Katerelos saved for each of the days throughout the periods January 1, 1987, through September 14, 1987, and May 30, 1990, through December 31, 1990, during which the time of day of each transaction was printed on that tape, Mr. Phielix created a third computerized *384 summary (third computerized summary). That summary showed for each day during those periods on which NDV was open for business: (1) The respective times of the first and last transactions recorded on the cash register tape that Mr. Katerelos saved, (2) the number of hours between those transactions, and (3) the number of missing transactions between the first numbered transaction on the cash register tape for that day and the last numbered transaction on the cash register tape for the preceding day. The third computerized summary showed that, for many days during the periods January 1, 1987, through September 14, 1987, and May 30, 1990, through December 31, 1990, the times printed on the cash register tapes that Mr. Katerelos saved indicated that the first transaction was entered before or shortly after 7 a.m. and the last transaction was entered just prior to or after 10 p.m. 26 However, that summary further showed that, for many other days during those same periods, the times printed on the cash register tapes that Mr. Katerelos saved indicated that the first transaction was not entered until substantially after NDV's normal opening time of 7 a.m. or the last transaction was entered*385 substantially earlier than its normal closing time of 10 p.m. In addition, the third computerized summary showed that a number of transactions were often missing and did not appear on any of the cash register tapes saved by Mr. Katerelos for days on which the times printed on the cash register tapes saved by Mr. Katerelos indicated that NDV either opened late or closed early. *386 To illustrate, the first transaction recorded on the cash register tape that Mr. Katerelos saved for January 4, 1987, a Sunday, was not entered until 4:39 p.m. and the last transaction recorded on that tape was entered at 9:18 p.m. Thus, according to the cash register tape that Mr. Katerelos saved for that day, the sales transactions that NDV had on that day spanned only 4 hours and 39 minutes. However, according to the cash register tapes that he saved for that day and the immediately preceding day, 79 transactions totaling $ 808.63 were entered into the cash register between 8:42 p.m. on January 3, 1987, when the Z or reset mode was activated, and 4:39 p.m. on January 4, 1987, when the first transaction recorded on the cash register tape that Mr. Katerelos saved for that day was entered into the cash register. Those 79 missing transactions did not appear on the cash register tapes Mr. Katerelos saved for either of those two days. By way of further illustration, the first transaction shown on the cash register tape that Mr. Katerelos saved for Valentine's Day 1987, a Saturday, was recorded at 4:49 p.m. However, according to the cash register tapes that he saved for that day and*387 the immediately preceding day, 70 transactions totaling $ 457.28 were entered into the cash register throughout the period that began with the last transaction entered into the cash register on the day before Valentine's Day and that ended with the first transaction entered on Valentine's Day. According to the cash register tapes that Mr. Katerelos saved, the last transaction on every Sunday during the period June through August 1990 was entered several hours prior to 10 p.m., and there were missing transactions that had been entered between the last transaction shown on those tapes for each of those Sundays and the first transaction shown on those tapes for each of the succeeding days during that same period. Mr. Phielix first raised with Mr. Katerelos his findings that petitioners omitted gross receipts of NDV from their returns for 1987 through 1989 when he met with petitioners on September 6, 1990. After that meeting, very few transactions were missing from the cash register tape that Mr. Katerelos saved for each of the days after September 6, 1990, through 1991, and the total sales according to the Z reading on that tape for each such day generally corresponded to each of the*388 Z reading differences in grand totals that Mr. Phielix calculated. Tax Returns Petitioners filed returns for the years 1987 through 1989 that were prepared by Mr. Berlat. 27 In those returns, petitioners reported gross receipts and net profit of NDV and total gross income, as follows: Gross Receipts Net Profit Year of NDV of NDV Total Gross Income [1987] $ 323,509 $ 12,584 $ 21,094 [1988] 323,037 10,378 8,862 [1989] 350,540 20,262 20,372 During 1990, petitioners filed amended returns for each of the years 1987 through 1989 that were prepared by petitioners' accountant, Mr. Bailey, a certified public accountant (C.P.A.) whom petitioners hired as their accountant during respondent's examination of their returns for 1987 through 1989 because*389 Mr. Berlat was not a C.P.A. Petitioners indicated in each of those amended returns that they had additional gross receipts of $ 5,200 attributable to "Cash - Register". Mr. Bailey prepared the amended returns for the years 1987 through 1989 after he had advised Mr. Katerelos that Mr. Phielix had informed him that petitioners had underreported the gross receipts of NDV for those years. Mr. Bailey prepared petitioners' returns for 1990, 1991, 1992, and 1993. In those returns, petitioners reported gross receipts and net profit of NDV and total gross income, as follows: Gross Receipts Net Profit Year of NDV of NDV Total Gross Income [1990] $ 372,332 $ 25,647 $ 22,626 [1991] 365,611 21,076 18,861 [1992] 396,801 38,694 42,614 [1993] 439,578 65,425 65,425 Petitioners' Expenditures and Other Relevant Admissions During the Years at Issue In a residential loan application signed by petitioners during December 1987, petitioners indicated that their monthly gross income was $ 5,000. Petitioners' monthly mortgage payment after securing that loan was $ 887.27. In an application for casualty insurance for NDV that Mr. Katerelos signed and that is dated*390 April 8, 1988, he stated that NDV had annual sales of $ 360,000. During 1987, petitioners made $ 29,000 of improvements to the residence that they had purchased in 1986. A list of those improvements appeared in NDV's gross receipts journal for 1987 at the time in 1990 that that journal was provided to Mr. Phielix during respondent's examination of petitioners' returns for 1987 through 1989. One week after he returned that journal to petitioners and confronted them with the list of improvements, Mr. Phielix reviewed the journal again and noticed that that list was missing. Mr. Katerelos had torn that list out of the gross receipts journal for 1987 sometime during that intervening one-week period. During 1987, petitioners and their family traveled to Disneyland and Las Vegas. During 1988, petitioners purchased airplane tickets for over $ 4,900 for Ms. Katerelos and other family members to fly to Greece. During August 1988, Mr. Katerelos purchased a 1987 Cadillac. Under the terms of a loan from Valley National Bank to finance that purchase, he made a downpayment of $ 4,500 and monthly payments of $ 330. All the payments for the purchase of that car, including the downpayment, were*391 made from NDV's checking account at Valley National Bank. Prior to purchasing the Cadillac, Mr. Katerelos owned a 1974 Lincoln. In the application for insurance that Mr. Katerelos signed in May 1988 for the Lincoln, the box marked "no" was checked in response to the question in that application as to whether that vehicle was used for business purposes. During the years 1988 through 1990, in addition to petitioners, NDV had one full-time employee. During 1988, Mr. Katerelos submitted an application for group health insurance for employees of NDV through Central Reserve Life of North America. The policy issued covered petitioners and their children, and not NDV's only other full-time employee. During 1989, Mr. Katerelos purchased a 14-carat gold ring for $ 1,467, and petitioners installed a swimming pool for which they paid Wilson Swimming Pools more than $ 14,000 and a patio and barbeque for which they paid more than $ 3,300. During January 1989, Mr. Katerelos purchased a 1985 Honda Accord for his daughter Anneta. In the application for insurance for that car, dated January 16, 1989, Mr. Katerelos was listed as the driver of the car for 60 percent of the time, Ms. Katerelos was*392 listed as the driver for 20 percent of the time, and his daughter Anneta was listed as the driver for 20 percent of the time. In fact, that car was primarily driven by petitioner's daughter Anneta after it was purchased. Petitioners borrowed $ 9,000 in May 1989 from Valley National Bank. They paid off that loan by making a principal payment of $ 6,000 during August 1989 and two principal payments totaling $ 3,000 during November 1989. In an application for insurance for a 1987 Chevrolet S-10 Blazer, dated June 10, 1991, petitioners were listed as the owners of the vehicle and Mr. Katerelos was listed as the driver of that vehicle for 30 percent of the time, Ms. Katerelos was listed as the driver for 50 percent of the time, and petitioners' son John was listed as the driver for 20 percent of the time. In fact, that vehicle was primarily driven by petitioners' son John after it was purchased. From 1988 to 1992, Mr. Katerelos was a 50-percent owner of Kefalonia, Inc. (Kefalonia) that operated a restaurant called Dolce Vita East. Mr. Katerelos' original investment in Kefalonia was $ 4,200, which he recovered at the rate of $ 175 per month. In 1992, Mr. Katerelos sold his interest in*393 Kefalonia for $ 20,000. Automobile Deductions Claimed During the years at issue, petitioners used Valley National Bank for their personal and business accounts. One branch of that bank was located approximately .7 miles from NDV, and another branch was located approximately 10 miles from NDV. During at least certain of the years at issue, Mr. Katerelos shopped at a Price Club for supplies for NDV and always wrote checks for those supplies. During at least part of the years at issue, a Price Club was located approximately 10 miles from NDV. In 1988, a second Price Club opened for business only .2 miles from NDV. After the second Price Club opened, Mr. Katerelos wrote checks only to that store to purchase supplies for NDV. Three weeks prior to the trial herein, Mr. Katerelos prepared a summary of his claimed automobile mileage for business purposes for 1987 through 1991. All of the miles listed in that summary were estimated, and he did not keep contemporaneous records of the actual miles traveled for alleged business purposes. Former Employees of NDV From 1988 to 1991, Elizabeth Solle (Ms. Solle) worked as a server at NDV. She generally worked about 20-25 hours per*394 week during the evening shift and usually did not work on Tuesday or Wednesday evenings. During certain periods of her employment, Ms. Solle did not work on Sundays or Mondays. Mr. Katerelos terminated her employment at NDV in 1991. While working at NDV, Ms. Solle never saw anyone being trained on the cash register using hypothetical sales or before or after normal business hours. From approximately 1988 through 1993, June Yubeta (Ms. Yubeta) worked as a hostess at NDV. She generally worked approximately 20 hours per week during the evenings and on Sunday mornings. Ms. Yubeta worked every Sunday while she was employed at NDV, except possibly one or two Sundays on which she was sick. As a hostess, Ms. Yubeta was responsible for operating the cash register. While working at NDV, Ms. Yubeta never entered a hypothetical sale into the cash register for training purposes and never saw anyone being trained on the cash register before or after normal business hours. From August 1988 to December 1989, Carmine Solle (Mr. Solle) worked at NDV as a dishwasher. He worked five days a week from 5 p.m. until 10 or 10:30 p.m. and generally had Tuesday and Wednesday evenings off. While working at*395 NDV, Mr. Solle never saw anyone being trained on the cash register after business hours. Judith DiIoli (Ms. DiIoli) worked at NDV as a hostess for approximately six months during the years at issue. She generally worked during the evenings from 3 p.m. to closing and on Saturday mornings. As a hostess, Ms. DiIoli was responsible for operating the cash register. While working at NDV, she entered actual, not hypothetical, sales into the cash register for purposes of her training and was trained during normal business hours. From 1987 until approximately 1991 or 1992, Margaret Rogers (Ms. Rogers) was employed at NDV as a server. She generally worked the breakfast and lunch shifts and usually did not work on Tuesdays or Wednesdays. As a server working the breakfast and lunch shifts, Ms. Rogers was responsible for operating the cash register. While working at NDV, she did not enter hypothetical sales into the cash register for the purpose of training. Ted Fisher Throughout all relevant periods, including the time of the trial herein, Ted Fisher (Mr. Fisher), who sells restaurant supplies, was a close, personal friend of Mr. Katerelos. During the years at issue, they saw each other*396 at least twice a week. Mr. Fisher did not notice any supplies on the premises of petitioners' house at the time they purchased it. OPINION Petitioners bear the burden of proving that the determinations in the notice are erroneous except those with respect to the additions to tax and penalties for fraud as to which respondent has the burden of proof that she must satisfy by clear and convincing evidence. Sec. 7454(a); Rule 142(a) and (b). Mr. Katerelos testified at trial. Based on the Court's opportunity to observe his demeanor and evaluate his credibility, we did not find him credible. In addition, we found that Mr. Katerelos had a selective memory and that he was evasive when answering questions posed by counsel for respondent. Under such circumstances, we are not required to, and generally do not, rely on the testimony of Mr. Katerelos. Geiger v. Commissioner, 440 F.2d 688, 689-690 (9th Cir. 1971), affg. per curiam T.C. Memo. 1969-159; Wood v. Commissioner, 338 F.2d 602, 605 (9th Cir. 1964), affg. 41 T.C. 593 (1964); Tokarski v. Commissioner, 87 T.C. 74, 77 (1986);*397 Hradesky v. Commissioner, 65 T.C. 87, 90 (1975), affd. per curiam 540 F.2d 821 (5th Cir. 1976). Issues As to Which Respondent Has the Burden of Proof Additions to Tax and Penalties for Fraud Respondent determined that Mr. Katerelos is liable for 1987 and 1988 for the additions to tax under section 6653(b)(1) and for 1989 and 1990 for the penalties under section 6663(a) on the portion of the underpayment attributable to the underreporting of certain gross receipts of NDV. 28*398 Respondent concedes that Ms. Katerelos is not liable for any of the years 1987 through 1990 for any additions to tax or penalties for fraud. 29 For 1987, section 6653(b)(1)(A) imposes*399 an addition to tax equal to 75 percent of the portion of an underpayment that is attributable to fraud and section 6653(b)(1)(B) imposes an addition to tax equal to 50 percent of the interest due on that portion of the underpayment. Section 6653(b)(1) for 1988 and section 6663(a) for 1989 and 1990 impose an addition to tax and a penalty, respectively, equal to 75 percent of the portion of an underpayment that is attributable to fraud. For purposes of the foregoing provisions, section 6653(b)(2) for 1987 and 1988 and section 6663(b) for 1989 and 1990 provide that if respondent establishes that any portion of an underpayment is due to fraud, the entire underpayment is treated as attributable to fraud, unless the taxpayer proves that some portion of the underpayment is not due to fraud. In order for the additions to tax and penalties for fraud to apply, respondent must prove by clear and convincing evidence that an underpayment exists and that some portion of such underpayment is due to fraud. Sec. 7454(a); Rule 142(b); Neidringhaus v. Commissioner, 99 T.C. 202, 210 (1992). To prove the existence of an underpayment, respondent may not rely on the taxpayer's*400 failure to carry his or her burden of proof with respect to the underlying deficiency. Parks v. Commissioner, 94 T.C. 654, 660-661 (1990); Petzoldt v. Commissioner, 92 T.C. 661, 700 (1989). Respondent must prove only that an underpayment exists, and not the precise amount of such underpayment. DiLeo v. Commissioner, 96 T.C. 858, 872 (1991), affd. 959 F.2d 16 (2d Cir. 1992); Petzoldt v. Commissioner, supra at 699. Underpayment of Tax Respondent contends that Mr. Katerelos underreported the gross receipts of NDV by $ 40,111, $ 65,009, $ 41,190, and $ 52,111 for the years 1987, 1988, 1989, and 1990, respectively, which resulted in an underpayment of income tax for each of those years. 30 To support that contention, respondent relies on various admissions by petitioners and the analyses by respondent of the cash register tapes that Mr. Katerelos saved for the years 1987 through 1990. We turn first to petitioners' admissions. *401 Petitioners admitted in the amended returns they filed during 1990 for each of the years 1987 through 1989 that they had underreported in the returns they originally filed for each of those years NDV's gross receipts for each such year. Each of those amended returns increased by $ 5,200 the amount of the gross receipts of NDV that petitioners had originally reported for each of the years 1987 through 1989. Although the filing of those amended returns is not evidence of fraudulent intent, it is evidence of an underpayment of tax for each of those years. See Badaracco v. Commissioner, 464 U.S. 386, 399 (1984); see also Old Mission Portland Cement Co. v. Commissioner, 69 F.2d 676, 680 (9th Cir. 1934), affg. in part and revg. in part 25 B.T.A. 305 (1932); Roche v. Commissioner, 63 F.2d 623, 624 (5th Cir. 1933), affg. 21 B.T.A. 1139 (1931). In addition to the amended returns petitioners filed for the years 1987 through 1989, on two other occasions, petitioners admitted having gross income and/or gross receipts for 1987 and 1988 in excess of*402 that reported in petitioners' return for each of those two years. In the residential loan application signed by petitioners during December 1987, petitioners indicated that their monthly gross income was $ 5,000, or approximately $ 60,000 per year. In their return for 1987, they reported gross income of only $ 21,094. Similarly, in an application for casualty insurance for NDV that Mr. Katerelos signed and that is dated April 8, 1988, he indicated that NDV had annual sales of $ 360,000. In their return for 1988, petitioners reported that NDV's gross receipts were only $ 323,037. 31 *403 Turning now to respondent's analyses of the cash register tapes that Mr. Katerelos saved for the years 1987 through 1990, Mr. Katerelos saved one cash register tape for each day throughout those years on which NDV was open for business. Those cash register tapes contained detailed information concerning certain transactions entered into the cash register, including (1) a nonresettable, consecutive transaction number for each transaction, (2) the transaction total of each transaction, (3) the date of each transaction, and (4) for certain periods during the years at issue, the time of each transaction. In addition, a Z reading on each cash register tape that Mr. Katerelos saved for the years at issue showed a summary of all the transactions entered into the cash register since the last Z reading, including the total sales since the last Z reading and the amount of the nonresettable grand total as of the time of the Z reading. 32 *404 Although Mr. Katerelos utilized the cash register tapes that he saved in order to calculate the amounts of gross receipts of NDV that he recorded in the gross receipts journals for 1987 through 1990 and that he reported in petitioners' returns for those years, he did not save the entire cash register tape generated by the cash register for each day on which NDV was open for business during those years. For certain days during the years 1987 through 1990, Mr. Katerelos activated the Z mode, thereby resetting the cash register, more than once a day. 33 However, the cash register tape that he saved for each such day showed only one Z reading, and it was that Z reading (adjusted as described above by any amount treated as an overring or cash shortage) from which he calculated the amount of gross receipts of NDV that he recorded in the gross receipts journals and reported in petitioners' returns for the years 1987 through 1990. Mr. Katerelos discarded the portions of the cash register tape for each such day that showed any other Z reading for that day. The missing portions of the cash register tapes that Mr. Katerelos discarded for 1987 through 1990 not only displayed Z readings that*405 Mr. Katerelos ignored in calculating the amounts of gross receipts of NDV that petitioners reported in their returns for those years, they also listed transactions that increased the grand total maintained by the cash register. By analyzing the cash register tapes that Mr. Katerelos saved for 1987 through 1990, respondent was able to ascertain the number of missing transactions that did not appear on any of the cash register tapes that Mr. Katerelos saved for those years and the Z reading differences in grand totals. Respondent determined that the amount of any such Z reading difference, less (1) the amount of gross receipts recorded by Mr. Katerelos in the gross receipts journals for a particular*406 day and (2) any adjustments for amounts treated as overrings or shortages of cash for that particular day, constituted the amount of daily gross receipts of NDV that Mr. Katerelos was required to, but did not, report in petitioners' returns for each of the years 1987 through 1990. 34 The amount of daily gross receipts of NDV that Mr. Katerelos had recorded in the gross receipts journals for a minimum of 650 of the 1,440 *407 days throughout the period 1987 through 1990 on which NDV was open for business was less than the amount of daily gross receipts of NDV that Mr. Phielix had calculated for each such day by using the Z reading difference in grand totals. Throughout the period 1987 through 1990, more than 11,400 of the approximately 140,000 transactions that were entered into the cash register during that period did not appear on any of the cash register tapes that Mr. Katerelos saved for those years. The total of the various amounts entered into the cash register for those missing transactions during the period that began on January 1, 1987, and that ended on December 31, 1990, was approximately $ 199,000. Mr. Katerelos does not dispute that portions of the cash register tapes are missing for the years 1987 through 1990 or that transactions were entered into the cash register during those years that do not appear on the cash register tapes that he saved. In fact, Mr. Katerelos admitted that he discarded (1) the portions of the cash register tapes that he did not utilize to calculate the amounts of gross receipts that he recorded in the gross receipts journals during the years 1987 through 1990 and*408 (2) the guest checks. It is his position that the missing transactions were nonsale transactions, such as corrections for amounts treated as overrings or transactions entered into the cash register for purposes of training employees on its use, that did not generate gross receipts of NDV that he failed to report in petitioners' returns for the years 1987 through 1990. To support that position, Mr. Katerelos relies on his own testimony. We did not find Mr. Katerelos' testimony that the missing transactions were nonsale transactions that did not constitute gross receipts of NDV for the years 1987 through 1990 to be credible. As for his testimony that the missing transactions resulted from the correction of overrings, the record establishes that Mr. Katerelos and/or other employees of NDV accounted for overring transactions that occurred on certain days during the years at issue by circling the overring transactions on the cash register tape for any such day and that Mr. Katerelos subtracted the transaction total of any such transaction from the total sales on the cash register tape that he saved for each such day in order to calculate the amount of NDV's gross receipts for that day*409 that he recorded in the gross receipts journal. We reject Mr. Katerelos' testimony that the missing transactions were overrings that were not accounted for by using the method described above. 35 As for Mr. Katerelos' testimony that he trained persons to operate the cash register by having them enter hypothetical sales into it, certain individuals who worked at NDV during the years at issue testified that they themselves never entered*410 any such hypothetical sales into the cash register. 36 Nor did they observe anyone entering such sales into the cash register. We reject Mr. Katerelos' testimony that the missing transactions were hypothetical sales entered into the cash register for purposes of training employees on its use. Mr. Katerelos also argues that respondent's analyses as shown in the second computerized summary created by Mr. Phielix are flawed because they show that NDV had high sales volumes on Mondays during 1987 through 1990, whereas, in fact, NDV had low sales volumes on Mondays during those years. We disagree. In determining the amount by which Mr. Katerelos underreported*411 the gross receipts of NDV for a particular day, which he displayed in the second computerized summary, Mr. Phielix assumed that the missing transactions were entered into the cash register on the last day of the multiday period in respect of which he was comparing the grand totals shown on the cash register tapes that Mr. Katerelos saved for the years at issue. Mr. Phielix made that assumption because although he was able to determine from examining those tapes that there were missing transactions, he was not able to determine from the cash register tapes whether those missing transactions were entered into the cash register on the last day of that period before the first transaction shown on that day's cash register tape was entered or on the first (or any other) day of that period after the last transaction shown on that day's cash register tape was entered. We agree with respondent that it is likely that the gross receipts of NDV generated by the missing transactions that Mr. Phielix assumed were attributable to Mondays during the years 1988 through 1990 were actually attributable to the preceding Sundays, days on which NDV generally had high sales volumes. An analysis of the *412 time at which the last transaction was entered into the cash register as shown on the cash register tapes that Mr. Katerelos saved for Sundays for the period June through August 1990 shows that on every Sunday during that period the last transaction entered into the cash register was entered several hours prior to 10 p.m. Respondent has also shown that, according to the cash register tapes that Mr. Katerelos saved for many of the days throughout the periods January 1, 1987, through September 14, 1987, and from May 30, 1990, through December 31, 1990, during which the cash register was programmed to print on the cash register tapes the time of day of every transaction using a clock, the first transaction at NDV was not entered into the cash register until substantially after NDV's normal opening time of 7 a.m. and/or the last transaction was entered into the cash register substantially earlier that NDV's normal closing time of 10 p.m. To illustrate, the first transaction recorded on the cash register tape for January 4, 1987, a Sunday, was not entered until 4:39 p.m., and the last transaction recorded on that tape was entered at 9:18 p.m. Thus, according to the cash register tape that*413 Mr. Katerelos saved for January 4, 1987, the sales transactions that NDV had on that day spanned only 4 hours and 39 minutes. However, 79 missing transactions totaling $ 808.63 were entered into the cash register between 8:42 p.m. on January 3, 1987, when the Z or reset mode was activated, and 4:39 p.m. on January 4, 1987, when the first transaction recorded on the cash register tape that Mr. Katerelos saved for that day was entered into the cash register. 37 Those 79 missing transactions do not appear on the cash register tapes that Mr. Katerelos saved for either of those two days. *414 By way of further illustration, the first transaction shown on the cash register tape for Valentine's Day 1987, a Saturday, was recorded at 4:49 p.m. However, 70 transactions totaling $ 457.28 were entered into the cash register throughout the period that began with the last transaction entered into the cash register on the day before Valentine's Day and that ended with the first transaction entered on Valentine's Day. Mr. Katerelos has not offered any credible explanation for the allegedly early closing or late opening of NDV shown on the cash register tapes that he saved for many days throughout the periods January 1, 1987, through September 14, 1987, and May 30, 1990, through December 31, 1990, during which the cash register printed the time of each transaction on the cash register tapes. Mr. Katerelos suggested during his testimony that he sometimes opened NDV late or closed it early if business were light or if one of his family members were sick. However, he could not explain why the restaurant would have opened late or closed early on any particular day during the years at issue or why NDV would have often opened late during 1987 on Sundays after breakfast and lunch, which*415 were among its busiest times. Moreover, Mr. Katerelos' testimony was contradicted by the former employees of NDV who testified that they did not remember any occasion on which NDV opened late or closed early, but that, if it had, that would have occurred only on very rare occasions. Respondent has also shown that petitioners made certain large expenditures during 1987, 1988, 1989, and 1990 that are inconsistent with the gross income reported by petitioners for those years. Mr. Katerelos alleges several sources for those expenditures other than his underreporting of the gross receipts of NDV. However, we did not find his testimony to be credible, and we are unwilling to rely on the testimony of Mr. Fisher, a close personal friend of Mr. Katerelos, who testified on his behalf about the alleged source of the funds for those expenditures. Based on our examination of the entire record before us, we find that respondent has established by clear and convincing evidence that there was an underpayment of petitioners' income tax for each of the years 1987 through 1990 as a result of Mr. Katerelos' underreporting certain gross receipts of NDV for each of those years. Fraudulent Intent *416 To prove that an underpayment is attributable to the fraudulent intent of a taxpayer, respondent must prove that the taxpayer intended to evade taxes that he or she believed to be owing by conduct intended to conceal, mislead, or otherwise prevent the collection of such taxes. Stoltzfus v. United States, 398 F.2d 1002, 1004 (3d Cir. 1968); Parks v. Commissioner, 94 T.C. at 661; see Laurins v. Commissioner, 889 F.2d 910, 913 (9th Cir. 1989), affg. Norman v. Commissioner, T.C. Memo. 1987-265. The existence of fraudulent intent is a question of fact to be resolved upon consideration of the entire record. Gajewski v. Commissioner, 67 T.C. 181, 199 (1976), affd. without published opinion 578 F.2d 1383 (8th Cir. 1978). Although fraudulent intent is never presumed, it may be established by circumstantial evidence. This is because direct proof of the taxpayer's intent is rarely available. Powell v. Granquist, 252 F.2d 56, 61 (9th Cir. 1958); Petzoldt v. Commissioner, 92 T.C. at 699;*417 Rowlee v. Commissioner, 80 T.C. 1111, 1123 (1983). The taxpayer's entire course of conduct may establish the requisite fraudulent intent. Otsuki v. Commissioner, 53 T.C. at 106. The Courts have identified a number of badges of fraud from which fraudulent intent may be inferred. Those badges include: (1) Substantial and consistent understatement of income, (2) inadequate records, (3) incomplete and erroneous information provided to tax return preparer or bookkeeper, (4) destruction of books and records, (5) implausible explanations of behavior, and (6) lack of credibility of the taxpayer's testimony. Bradford v. Commissioner, 796 F.2d 303, 307-308 (9th Cir. 1986), affg. T.C. Memo. 1984-601; Toussaint v. Commissioner, 743 F.2d 309, 312 (5th Cir. 1984), affg. T.C. Memo. 1984-25; Merritt v. Commissioner, 301 F.2d 484, 487 (5th Cir. 1962), affg. T.C. Memo. 1959-172; Powell v. Granquist, supra at 59; Parks v. Commissioner, 94 T.C. at 664-665.*418 A pattern of underreporting income over an extended period of time is indicative of fraud. Laurins v. Commissioner, supra at 913; Petzoldt v. Commissioner, supra at 700. However, the mere failure to report income is not sufficient to establish fraud. Petzoldt v. Commissioner, supra at 700. The record in this case establishes, inter alia, that (1) Mr. Katerelos substantially underreported the gross receipts of NDV for each of the years 1987 through 1990; (2) he activated the Z or reset mode on the cash register more than once on a number of days during each of those years, but reported the total sales from only one such Z reading on those days as the amount of gross receipts of NDV even though he knew he was required to report the transaction totals of all sales made by NDV as gross receipts for those years; (3) during each of the years 1987 through 1990, Mr. Katerelos discarded the portions of the cash register tapes that he did not utilize to calculate the amounts of gross receipts that he recorded in the gross receipts journal for each such year; (4) Mr. Katerelos discarded all of the guest*419 checks for the years 1987 through 1990; (5) Mr. Katerelos provided his accountant with only the gross receipts journals and the portions of the cash register tapes that he saved and utilized to calculate the amounts of gross receipts that he recorded therein; (6) the pattern of missing transactions repeated itself on a regular basis during each of the years 1987 through 1990 until September 1990 when Mr. Phielix first confronted petitioners concerning his analyses of the cash register tapes; (7) Mr. Katerelos provided implausible explanations as to the nature of the missing transactions and as to why the cash register tapes for certain days indicated NDV was not open during normal business hours; and (8) Mr. Katerelos' testimony at trial was not credible. Although the additions to tax and the penalties for fraud apply to the entire underpayment unless the taxpayer proves that some portion of the underpayment is not due to fraud, respondent determined, and does not argue otherwise herein, that only the portion of the underpayment for each of the years 1987 through 1990 that is attributable to the underreporting of certain gross receipts of NDV for each such year is due to fraud. Mr. *420 Katerelos has not shown either (1) that he underreported in petitioners' returns for the years 1987 through 1990 the gross receipts of NDV in any amounts other than the amounts by which respondent contends that he underreported those gross receipts or (2) that any portion of the underpayment for each of those years that is attributable to that underreporting is not due to fraud. Based on our examination of the entire record before us, we find that respondent has established by clear and convincing evidence that Mr. Katerelos intended to evade tax for each of the years 1987 through 1990 that he knew to be owing by conduct intended to conceal, mislead, or otherwise prevent the collection of such taxes. We find on the record before us that Mr. Katerelos is liable for 1987 and 1988 for the additions to tax for fraud under section 6653(b)(1) and for 1989 and 1990 for the penalties for fraud under section 6663(a) with respect to the entire portion of the underpayment for each of those years that is attributable to the underreporting of gross receipts of NDV. Issues as to Which Petitioners Have the Burden of Proof Underreporting of Gross Receipts of NDV Respondent determined*421 in the notice that petitioners underreported the gross receipts of NDV by $ 40,111, $ 65,102, $ 41,190, and $ 60,094 for the years 1987, 1988, 1989, and 1990, respectively. 38 On brief, respondent reduces the amounts of the gross receipts that she claims petitioners underreported for 1988 and 1990 to $ 65,009 and $ 52,111, respectively. We found above on the record before us that respondent proved by clear and convincing evidence that Mr. Katerelos underreported certain gross receipts of NDV in petitioners' returns for each of the years 1987 through 1990. A fortiori, we further find on that record that petitioners failed to sustain their burden of proving error in respondent's determination*422 that they underreported those gross receipts for each of those years. Accordingly, we sustain respondent's determination, as modified on brief, that petitioners underreported the gross receipts of NDV by $ 40,111, $ 65,009, $ 41,190, and $ 52,111 for the years 1987, 1988, 1989, and 1990, respectively. Automobile Expenses Petitioners argue that they are entitled to deductions under section 162(a) for each of the years at issue for expenses incurred with respect to the use of an automobile. Respondent determined in the notice that petitioners are not entitled to the automobile expense deductions that they claimed in their returns for 1987, 1989, 1990, and 1991 39 and argues that they are not entitled to the deductions for such expenses that they claimed in the amended return they filed for 1988. 40 Petitioners bear the burden of proof on this issue. *423 Section 162(a) generally allows a deduction for ordinary and necessary expenses paid or incurred during the taxable year in carrying on a trade or business. The determination of whether an expenditure satisfies the requirements for deductibility under section 162 is a question of fact. Commissioner v. Heininger, 320 U.S. 467, 475 (1943). In general, an expense is ordinary if it is considered "normal, usual, or customary" in the context of the particular business out of which it arose. Deputy v. du Pont, 308 U.S. 488, 495-496 (1940). Ordinarily, an expense is necessary if it is appropriate and helpful to the operation of the taxpayer's trade or business. Commissioner v. Tellier, 383 U.S. 687, 689 (1966); Carbine v. Commissioner, 83 T.C. 356, 363 (1984), affd. 777 F.2d 662 (11th Cir. 1985). To substantiate their claimed deductions for automobile expenses, petitioners rely on the testimony of Mr. Katerelos and a document he prepared three weeks prior to the trial herein that purports to reconstruct in very summary fashion the business*424 use of his automobile (summary). We are unwilling to rely on the testimony of Mr. Katerelos or on the document he prepared in anticipation of trial to prove either (1) that petitioners actually incurred such expenses or (2) that if such expenses were incurred, they were ordinary and necessary expenses with respect to petitioners' business, NDV. With respect to Mr. Katerelos' testimony regarding the automobile expenses at issue, in addition to our having found that his testimony is generally not credible, his testimony with respect to his use of an automobile for NDV's business is in conflict with an insurance application he signed in May 1988 for the Lincoln that he claims he used for business purposes throughout 1987 and during 1988 until he purchased a 1987 Cadillac. In that application, the box marked "no" was checked in response to a question as to whether the Lincoln was used for business purposes. Thus, according to that insurance form, Mr. Katerelos was not using the Lincoln for business purposes, and his testimony is inconsistent with that form. With respect to the summary of the automobile expenses at issue that Mr. Katerelos reconstructed, we did not find that self-serving*425 and uncorroborated reconstruction to be credible. In addition to our having found that Mr. Katerelos' testimony is generally not credible, we question how he could have remembered enough specific facts to prepare that summary when he testified on various occasions that he did not remember specific facts when questioned by counsel for respondent. 41 We also question the validity of the summary of automobile expenses that Mr. Katerelos prepared, since he claimed therein that (1) he traveled five times a week to a branch of his bank that was 10 miles away from NDV when another branch was located approximately .7 miles from NDV and (2) he continued to visit a Price Club once a week that was 10 miles from NDV even after a new Price Club opened in 1988 that was only .2 miles from NDV and to which he wrote all of the checks that he wrote to Price Club subsequent to its opening. *426 Based on our review of the entire record before us, we find that petitioners failed to prove that they are entitled to deductions under section 162(a) for automobile expenses for any of the years at issue. 42 We therefore sustain respondent's determinations disallowing those deductions for 1987, 1989, 1990, and 1991 and reject petitioners' contention that they are entitled to any such deduction for 1988. Depreciation Petitioners claimed depreciation deductions in their returns for each of the years at issue with respect to properties that they claim they placed in service for their business, NDV, during 1986 and that they claim had total bases of $ 76,052. Respondent determined in the notice that petitioners are not entitled to those depreciation deductions because they failed to establish (1) their bases in the properties with respect to which the depreciation deductions were claimed and (2) that*427 the properties with respect to which those deductions were claimed were depreciable assets. On brief, respondent argues that petitioners failed to establish what properties they placed in service during 1986 and their adjusted bases in any properties they did in fact place in service during that year. Petitioners bear the burden of proof on this issue. Section 167(a) allows a deduction for a reasonable allowance for the exhaustion, wear and tear of property used, inter alia, in a trade or business. Section 167(c)(1) provides that the basis on which the exhaustion, wear and tear is allowed is the adjusted basis determined under section 1011. Thus, to be entitled to a depreciation deduction, petitioners must establish their adjusted bases in the properties with respect to which they are claiming depreciation deductions for the years at issue. Prior to the trial herein, the parties stipulated that petitioners are entitled to depreciation deductions for the years at issue that are based on adjusted bases of a minimum of $ 22,102 with respect to properties placed in service during 1986. Petitioners argue that their bases in the properties placed in service during 1986 was greater than*428 the amount of $ 22,102 to which the parties stipulated. 43 The only evidence in the record regarding petitioners' claim is the testimony of Mr. Katerelos on which we are unwilling to rely. 44 Although petitioners admitted that they do not have receipts*429 to support their claim that they had bases in properties placed in service during 1986 in excess of the $ 22,102 to which the parties stipulated, they argue that they should be allowed to estimate from memory their bases in the properties that they placed in service during that year and that the Court should sustain that estimate. On the record before us, we decline to do so. Petitioners have failed to establish what properties they placed in service during 1986 or that their bases in whatever properties they did so place in service exceeded the $ 22,102 to which the parties stipulated, and the present record does not provide enough reliable information to enable us to make a reasonable estimate of the petitioners' bases in the properties with respect to which they claim depreciation deductions for the years at issue. See Coloman v. Commissioner, 540 F.2d 427, 431-432 (9th Cir. 1976), affg. T.C. Memo. 1974-78; Vanicek v. Commissioner, 85 T.C. 731, 742-743 (1985). Based on our review of the entire record before us, we find that petitioners failed to prove that their bases in the properties at *430 issue exceeded the $ 22,102 to which the parties stipulated. We therefore shall not allow petitioners to deduct for any of the years at issue any amount under section 167(a) in excess of that amount calculated by taking account of those stipulated bases. General Business Credit Carryforward In addition to claiming depreciation deductions for the years at issue with respect to the properties that they claim they placed in service during 1986, petitioners contend that they are entitled to a general business credit carryforward from 1986 with respect to those properties. Respondent determined that petitioners are not entitled to that credit because they failed to establish that they had a credit from 1986 to carry forward. Credits are a matter of legislative grace, and taxpayers must prove their entitlement to them. Rule 142(a); Segel v. Commissioner, 89 T.C. 816, 842 (1987). A taxpayer is required to maintain records as long as the contents of those records may become material in the administration of any internal revenue law. Sec. 1.6001-1(e), Income Tax Regs. If a credit affects the calculation of tax due for an open year, respondent is entitled*431 to recalculate the amount of credits earned and/or used for a particular year even if the period of limitations has expired for that year. Hill v. Commissioner, 95 T.C. 437, 445-446 (1990); Mennuto v. Commissioner, 56 T.C. 910, 923 (1971). Petitioners have failed to establish that they had a general business credit from 1986 to carry forward. For 1986, section 38(a)(2) allows a credit for, inter alia, the amount of the current year business credit. Section 38(b) defines the term "current year business credit" to include the investment credit determined under section 46(a). Sec. 38(b)(1). Section 46 defines the amount of the investment credit to include the regular percentage of the qualified investment determined under section 46(c). Sec. 46(a)(1). Section 46(c)(1) provides that the term "qualified investment" means with respect to any taxable year the aggregate of the applicable percentage of the basis of each new section 38 property, as defined in section 48(b), placed in service by the taxpayer during such year plus the applicable percentage of the cost of each used section 38 property, as defined in section 48(c)(1), *432 placed in service by the taxpayer during such year. Section 38 property is defined to include "tangible personal property (other than an air conditioning or heating unit)". Sec. 48(a)(1)(A). The term "section 38 property" "includes only recovery property (within the meaning of section 168 * * *) and any other property with respect to which depreciation * * * is allowable and having a useful life * * * of 3 years or more." Sec. 48(a). Petitioners have not established that (1) they placed section 38 property in service during 1986 or (2) either (a) their basis in new section 38 property or (b) the cost of used section 38 property. 45 *433 In addition, section 49(a) disallows the regular percentage of the qualified investment determined under section 46(c) for property placed in service after December 31, 1985. Section 49(b)(1) provides an exception to that disallowance for transition property as defined in section 49(e). Section 49(e)(1) defines transition property to mean any property placed in service after December 31, 1985, and to which the amendments made by section 201 of the Tax Reform Act of 1986 do not apply, except that in making such determination-- * * * * (B) sections 203(b)(1) and 204(a)(3) of such Act shall be applied by substituting "December 31, 1985" for "March 1, 1986", Section 203(b)(1) of the Tax Reform Act of 1986, Pub. L. 99-514, 100 Stat. 2143-2144, provides in pertinent part: (1) IN GENERAL.--The amendments made by section 201 shall not apply to-- (A) any property which is constructed, reconstructed, or acquired by the taxpayer pursuant to a written contract which was binding on March 1, 1986, (We shall hereinafter refer to the foregoing exception in section 203(b)(1)(A) of the Tax Reform Act of 1986, 100 Stat. 2144, as the binding contract exception.) The conference*434 committee report addressing the binding contract exception states that the binding contract exception "applies only to contracts in which the construction, reconstruction, erection, or acquisition of property is itself the subject matter of the contract." H. Conf. Rept. 99-841, at II-55 (1986), 1986-3 C.B. (Vol. 4) 55. Thus, in order for property to be transition property within the meaning of section 49(e)(1), it must be constructed, reconstructed, or acquired by the taxpayer pursuant to a written contract (1) that was binding on December 31, 1985, and (2) the subject matter of which was the construction, reconstruction, or acquisition of such property. During December 1985, petitioners executed a lease for the premises at which NDV was located. That lease provided that the leased premises were to be used to operate a restaurant. Petitioners argue that, in order to operate the leased premises as a restaurant, they were required to purchase property for use at the leased premises and that therefore the binding contract exception applies to property that they purchased for use at NDV. We disagree. The subject of the lease executed by petitioners in December*435 1985 was the use of the premises at which NDV was located, and not the construction, reconstruction, or acquisition of property for use at that premises. Consequently, the binding contract exception does not apply to the lease executed by petitioners for the premises at which NDV was located. On the instant record, we find that petitioners failed to prove that they are entitled to a general business credit carryforward from 1986. We therefore sustain respondent's determination on that issue. To reflect the foregoing and the concessions of the parties, Decision will be entered under Rule 155. Footnotes 1. All section references are to the Internal Revenue Code in effect for the years at issue. All Rule references are to the Tax Court Rules of Practice and Procedure.↩ 2. Petitioners' opening brief did not contain all of the findings of fact that petitioners wanted to propose and that they were required to set forth in that brief, as required by Rule 151(e)(3). Instead, in violation of that Rule, their answering brief contained additional proposed findings of fact that the Court did not find to be "alternative proposed findings of facts" as permitted by Rule 151(e)(3).↩ 3. It is unclear from the record who trained other individuals who operated the cash register during the years at issue.↩ 4. It is unclear from the record how often Mr. Katerelos entered sales into the cash register during the years at issue.↩ 5. Unless otherwise indicated, our discussion of operation of the cash register applies to all of the years at issue.↩ 6. For example, if the operator were entering a lunch sale into the cash register, that person entered the amount of the transaction followed by the lunch department button.↩ 7. It is unclear whether other operators of the cash register during the years at issue distinguished between cash and credit sales.↩ 8. That information could also have been accessed by activating the X or read mode on the cash register.↩ 9. The cash register tape is one of two printed tapes that the cash register produced. The other tape is designed to provide receipts to customers and is not relevant herein.↩ 10. Activation of the X or read mode printed a reading on the cash register tape similar to the Z reading; however, unlike the Z or reset mode that reset certain amounts to zero when it was activated, activation of the X mode did not reset any amounts. For certain days throughout the years at issue, Mr. Katerelos activated the X mode on the cash register at approximately 5 p.m. in order to subtotal and display on the cash register tape for each such day the amount of gross receipts through breakfast and lunch.↩ 11. The grand total was reset to zero on May 30, 1990. As discussed below, the only way to reset the grand total to zero was to cause the capacitors to drain by unsoldering and taking the battery out of circuit. The record does not provide any explanation as to why the battery was unsoldered and taken out of circuit as of May 30, 1990, and neither party has suggested that the resetting of the grand total as of May 30, 1990, should affect our findings or conclusions herein.↩ 12. When the grand total was reset to zero as of May 30, 1990, the transaction number was also reset to zero.↩ 13. As discussed below, the cash register tapes that Mr. Katerelos saved for certain days during the years at issue showed two Z readings.↩ 14. In addition, the transaction number was increased by one as a result of any overring transaction, and the grand total as of that day was increased by the transaction total of any overring transaction. Although Mr. Katerelos did not use the grand totals to determine the amount of NDV's gross receipts that he recorded for a particular day in the gross receipts journals, as discussed below, respondent's revenue agent used the grand totals to determine the amount of NDV's gross receipts for the years at issue, and, in calculating those gross receipts, he made appropriate adjustments for any transactions that were or were treated as overrings.↩ 15. It usually took Mr. Katerelos five to ten, and not more than 30 minutes, to reenter those transactions.↩ 16. Mr. Katerelos treated the total sales from the first Z reading as if it had been an overring (even if it was not) in that he did not include it in the amount of NDV's gross receipts that he recorded for that day in the gross receipts journal.↩ 17. Mr. Katerelos employed Mr. Berlat as the accountant for NDV when Mr. Berlat's father, who had been performing that function, died. Although it is unclear from the record when Mr. Berlat's father died, it appears that he may have passed away during 1987.↩ 18. Following the close of each month during each of the years at issue, Mr. Berlat visited the premises of NDV, obtained the gross receipts journal from Mr. Katerelos, and calculated from that journal the amount of NDV's gross receipts for the month just ended.↩ 19. The missing transactions and the missing transaction numbers also included Z readings. They also may have included transactions attributable to pressing the no-sale button in order to open the cash register. Neither of those types of transactions increased the grand total maintained by the cash register.↩ 20. If the cash register tape that Mr. Katerelos saved contained two Z readings, Mr. Phielix used the data displayed by the second Z reading when he created his handwritten summary.↩ 21. The Z reading difference in grand totals included a difference in grand totals according to the Z readings on the cash register tapes that Mr. Katerelos saved where, because Mr. Phielix was unable to read the respective grand total displayed on those tapes for certain days during the years at issue, he used the grand total that was legibly displayed on the cash register tape that Mr. Katerelos saved for the day next succeeding the day for which the cash register tape that Mr. Katerelos saved did not legibly display a grand total. Under this approach, instead of calculating the Z reading difference in grand totals for a two-day period, Mr. Phielix computed the Z reading difference in grand totals for a period consisting of more than two days.↩ 22. Mr. Phielix did not transfer into the computer the information for 1991 that was contained in his handwritten summary. That was because the amount of additional tax that Mr. Phielix had calculated that petitioners owed for that year as a result of the underreporting of NDV's gross receipts for that year appeared to him to be de minimis. Respondent did not determine in the notice of deficiency (notice) that petitioners were liable under sec. 6663(a)↩ for the penalty for fraud for any portion of the underpayment for 1991. 23. Neither party advances any argument that our findings or conclusions herein should be affected by any errors occurring during the transfer into the computer of data by Mr. Phielix from the cash register tapes that Mr. Katerelos saved.↩ 24. To illustrate, the cash register tape that Mr. Katerelos saved for Aug. 30, 1989, displayed two Z readings. That tape showed that (1) a series of transactions, the last of which included an error, had been entered into the cash register before the Z mode generating the first Z reading had been activated; (2) after the first Z reading was generated, the transaction that included the error was correctly entered into the cash register and the amount of each transaction except the erroneous transaction was reentered into the cash register as part of a single transaction that was assigned a single transaction number; (3) after those transactions were reentered into the cash register, a series of additional transactions was entered into the cash register; and (4) the Z mode was activated again, thereby generating the second Z reading displayed at the end of that tape. Mr. Katerelos recorded in the gross receipts journal for 1987 the total sales according to that second Z reading as the amount of gross receipts of NDV for Aug. 30, 1989. In determining the amount of gross receipts of NDV that Mr. Katerelos was required to report for that day, Mr. Phielix treated the total sales according to the first Z reading on the cash register tape that Mr. Katerelos saved for Aug. 30, 1989, as an overring and subtracted that amount from the Z reading difference in grand totals that he had calculated by using the second Z reading for Aug. 30, 1989. As a result, Mr. Phielix determined that petitioners had not underreported NDV's gross receipts for Aug. 30, 1989.↩ 25. When the Z reading difference in grand totals was for a period of more than two days, Mr. Phielix reduced the Z reading difference for that multiday period (adjusted as described above by any amount treated as an overring or cash shortage) by the sum of the amount of daily gross receipts that Mr. Katerelos had recorded in the gross receipts journals for each of the days during that period.↩ 26. It is not possible to determine from the cash register tape that Mr. Katerelos saved for a particular day whether the times printed thereon were the actual times of day at which transactions were recorded. This is because the clock could have been set incorrectly for that day. However, even if the clock had been set incorrectly for a particular day, all of the times of the various transactions that were printed on the cash register tape that Mr. Katerelos saved for that day would have reflected the same error. For example, if the clock for a particular day had been set an hour ahead of the actual time, the time of day for each transaction that was printed on the cash register tape that Mr. Katerelos saved for that day would have been an hour ahead. Nor would the number of hours between the first and last transactions for a particular day have been affected if the clock had been set incorrectly. In addition, unless the clock had been reset, the clock would have been incorrect by the same amount from one day to the next day.↩ 27. Both parties acknowledge that Mr. Berlat made certain errors in preparing petitioners' returns for the years 1987 through 1989. However, neither party has suggested that those errors should affect our findings or conclusions herein.↩ 28. Respondent determined in the notice that $ 18,151, $ 10,918, and $ 21,997 of the total underpayment for each of the years 1988 through 1990, respectively, was due to fraud. Respondent did not determine that petitioners are liable for the additions to tax and penalties for fraud on that portion of the underpayment for each of those years that was not attributable to the underreporting of certain gross receipts of NDV.↩ 29. Petitioners argue that, because fraud cannot be imputed from one spouse to another and because respondent concedes that Ms. Katerelos is not liable for any additions to tax or penalties for fraud for any of the years 1987 through 1990, respondent has not carried her burden of proving fraud. That respondent concedes that Ms. Katerelos is not liable for fraud does not relieve Mr. Katerelos from the additions to tax or penalties for fraud. That petitioners filed a joint return is not a defense to the imposition of such additions and penalties against only one spouse. See, e.g., Otsuki v. Commissioner, 53 T.C. 96, 112-113 (1969). Even the case cited by petitioners as authority for the proposition that fraud cannot be imputed from one spouse to another imposed the penalty for fraud against one spouse even though respondent failed to prove fraud against the other spouse. See Fry v. Commissioner, T.C. Memo. 1991-51, affd. without published opinion 8 F.3d 26↩ (9th Cir. 1993). 30. Respondent originally determined that petitioners underreported the gross receipts of NDV by $ 65,102 and $ 60,094 for 1988 and 1990, respectively; however, on brief, respondent reduces the amounts of the additional gross receipts that she claims were underreported for those years to $ 65,009 and $ 52,111, respectively.↩ 31. The $ 21,094 that petitioners reported as gross income in their return for 1987 plus the additional gross receipts of NDV of $ 40,111 that respondent determined, and that we sustain below, that petitioners failed to report for that year equals approximately $ 60,000 in gross income for 1987. The $ 323,509 that petitioners reported as gross receipts of NDV in their return for 1987 plus the additional gross receipts of NDV of $ 40,111 that respondent determined, and that we sustain below, that petitioners failed to report for that year equals approximately $ 360,000 of gross receipts for 1987. On the record before us, we believe that, when petitioners believed it was to their advantage, they admitted having additional gross income and/or additional gross receipts of NDV in almost the exact amount determined by respondent for 1987.↩ 32. Activation of the Z or reset mode also caused the cash register to reset all amounts (other than certain nonresettable amounts) to zero.↩ 33. Unless otherwise indicated, our discussion hereinafter concerning days on which Mr. Katerelos activated the Z or reset mode more than once a day does not pertain to any day on which Mr. Katerelos activated the Z mode only twice during the day and the cash register tape that he saved for that day showed both Z readings.↩ 34. The amounts that respondent determined constituted additional gross receipts of NDV that Mr. Katerelos did not report in petitioners' returns for 1987 through 1990 included sales tax that petitioners collected on sales by NDV. Petitioners do not argue that respondent's determinations of the amounts of their underpayments for those years should be reduced by the amounts of sales tax included in the unreported gross receipts determined by respondent, presumably because petitioners did not pay any sales tax that they collected on unreported transactions to the appropriate local tax authorities.↩ 35. We also reject Mr. Katerelos' testimony that the missing transactions were attributable to his having occasionally accounted during 1987 through 1990 for errors (that may or may not have been overrings) in entering transactions into the cash register by using the Z or reset mode after an error occurred, correctly reentering into the cash register each transaction for a given day that had been entered prior to the first Z reading, and using the Z or reset mode again in order to calculate the amount of NDV's gross receipts for that day that he recorded in the gross receipts journals.↩ 36. Although Ms. Yubeta initially testified that she was trained using hypothetical sales, we question that testimony. This is because Ms. Yubeta subsequently admitted during her testimony that she had never entered a hypothetical sale into the cash register and that she did not want to say anything to hurt Mr. Katerelos because she thought he was a "good employer".↩ 37. Because the last transaction recorded on the cash register tape for Jan. 3, 1987, was entered into the cash register at 8:42 p.m., it is possible that some of the 79 missing transactions for the two-day period Jan. 3-4, 1987, were entered into the cash register after 8:42 p.m., prior to the close of business on Jan. 3, 1987, and some were entered into the cash register on Jan. 4, 1987, prior to 4:39 p.m. at which time the first transaction appears on the cash register tape that Mr. Katerelos saved for that day. However, it does not matter on which of those two days the missing transactions were entered into the cash register. This is because, regardless on which of those two days those transactions were entered, the grand total maintained by the cash register was increased by the transaction total of each of those transactions.↩ 38. Although respondent determined in the notice that petitioners underreported the gross receipts of NDV by $ 4,714 for 1991, the parties stipulated that petitioners' gross receipts for 1991 should be increased by only $ 999.49. Thus, we need not address whether petitioners underreported any of NDV's gross receipts for that year.↩ 39. Petitioners did not claim any automobile expense deductions in their return for 1988.↩ 40. Respondent made no determination with respect to that amended return.↩ 41. For example, when asked whether construction on the swimming pool that petitioners had installed was started in May 1989, Mr. Katerelos responded, "I don't remember dates." Similarly, when asked whether he installed a $ 3,500 barbeque and patio, he responded, "I don't recall amounts."↩ 42. We note that petitioners also failed to prove that they satisfy the requirements imposed by sec. 274(d)(4).↩ 43. It is unclear whether petitioners are arguing that they placed properties into service during 1986 in addition to those properties that the parties stipulated had minimum bases of $ 22,102 or that their bases in the latter properties exceeded $ 22,102.↩ 44. Petitioners attempted to call Mr. Fisher to testify as an expert witness on their behalf for purposes of valuing the properties used at NDV. As required by Rule 143(f), petitioners submitted a purported expert report prepared by Mr. Fisher before trial. Respondent filed a motion to exclude that report and any expert testimony by Mr. Fisher. The Court granted respondent's motion and did not admit the purported expert report of Mr. Fisher into evidence and did not allow him to testify as an expert witness.↩ 45. Although it is possible that certain of the properties that the parties stipulated had minimum bases of $ 22,102 and that petitioners placed in service during 1986 do, in fact, fit within the definition of sec. 38 property, the record before us does not permit us to make such a determination. For example, petitioners have not even established what properties are covered by that stipulation, let alone whether such properties, whatever they are, constitute sec. 38 property.↩